Genetic improvement of meat animals has been mainly obtained through the selection of production traits related to fattening efficiency and retail product at slaughter. Up to know, animals have been selected on live growth and also, according to the species, more and more on muscle growth capacity, the main biological component of production. Only the technological quality of pork meat is presently taken into account in genetic improvement schemes owing to its economic importance, the discovery of major genes on meat defaults (HAL and RN) and the predictive measure of pH, lightness and water loss in slaughterhouses. Genotyping for the HAL and RN default is currently used for selecting breeding stocks. The measures of pH, lightness and water loss are included in synthetic indexes in order to maintain the technological quality level while improving lean growth. Poultry meat is being more and more used in industrial processes. Similarly to pork, the measures of pH, lightness and water loss are genetically linked to cooking loss. Genetic variability and correlations are even higher than in pork. However, these measures are genetically independent of production traits in poultry, while they are slightly unfavourably related in pork. Presently the selection for sensory quality attributes is not possible due to the lack of predictive measures in slaughterhouses. A genetic variability has, however, been estimated in the different species. Current studies are looking for the estimation of genetic relationships between production and quality traits. In poultry and rabbits, selection for growth has no impact on quality if the animals are slaughtered at the same age. However a reduction of the slaughter age of poultry as a consequence of selection for growth rate has been shown to improve tenderness and to decrease flavour correlatively to a lower maturity. In pork, genetic lean growth is markedly unfavourably related to the sensory quality. A degradation of the latter may be avoided if the intramuscular lipid content could be selected. In beef, most studies originate from North America or Australia with different types of cattle and in different consuming conditions than those found in France : young fattened steers versus older and leaner young bulls and higher cooking temperatures. These study results are therefore not useful for the French production systems and consuming conditions. French studies show that a selection for muscle growth capacity should be beneficial for tenderness, but detrimental for flavour. Similarly to pork, a selection for intramuscular lipids will counteract that negative effect. In both species, simultaneous improvement of muscle growth and intramuscular lipids is difficult due to the genetic antagonism between both traits and more certainly due to the difficulty to have an accurate, non destructive, predictive measure of lipids in these very lean animals. Due to the lack of an available predictive measure of sensory quality in slaughterhouses, current studies have been aimed at the research of polymorphic genes involved in meat quality that can be selected through genotyping. QTL detection programmes have been conducted and complemented with positional cloning and candidate gene approaches in order to find the causal mutations. The first genes suspected to influence meat quality have been commercialised by private enterprises. The segregation and the impact of the suspected genes have however to be validated within breed for selection purpose.